Figure 1.
Histological and functional analysis of PARP1 KO retina.
Haematoxylin/eosin staining at PN30 revealed normal morphology and layering of wt (A) retina, while in the rd1 situation (B) the ONL had almost completely disappeared. In contrast, PARP1 KO retinae (C) appeared essentially normal, although direct comparisons with wt showed lower PARP1 KO values for ONL thickness and number of photoreceptor rows (quantification in D). SD-OCT in vivo imaging of wt (E), rd1 (F), and PARP1 KO (G) retinae showed a similar picture, with PARP1 KO retina appearing slightly thinner than its wt counterpart. Absence of the 116 kDa PARP1 protein was confirmed using western blot (H). In spite of the subtle morphological changes seen in PARP1 KO, functional in vivo analysis using ERG under both scotopic and photopic conditions in 5 weeks old animals did not detect any differences between PARP1 KO (red traces) and wt control (black traces). In rd1 animals (blue traces), however, retinal function was essentially abolished. Representative single flash ERG recordings from dark-adapted (top) and light-adapted (bottom) states are shown in (I), while a statistical evaluation (box-and-whisker plot) of dark-adapted (scotopic; SC) and light-adapted (photopic; PH) single flash ERG b-wave amplitudes in wt, rd1, and PARP1 KO mice is shown in (J). Boxes indicate the 25% and 75% quantile range, whiskers indicate the 5% and 95% quantiles, and solid lines connect the medians of the data. For each of the different experimental investigations, n = 3–4 animals from each genotype were used and analyzed independently. Error bars in (D) represent SEM. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; IS, inner segment; OS, outer segment; RPE, retinal pigment epithelium.
Figure 2.
cGMP, PARP activity and TUNEL in wt, rd1, and PARP1 KO retina.
At P11, immunoreactivity for cGMP was essentially absent in both wt and PARP1 KO retina, while in the rd1 ONL a large number of photoreceptor cell bodies, neurites and segments were stained (A–C). The in situ PARP activity assay (D–F) and accumulation of PAR (G–I) as an indirect marker for PARP activity labeled photoreceptor nuclei only in the rd1 situation but not in wt nor PARP1 KO retina. The bar graphs display the quantification of the percentages of ONL cells positive for cGMP and TUNEL (M), and PARP activity and PAR accumulation (N), respectively. Strong accumulation of PAR in rd1 retina was confirmed using western blot (O). The TUNEL assay for dying cells identified large numbers of cells only in the rd1 ONL (J–L), Retinae from n = 3–6 animals were used for each analysis and genotype. Error bars represent SEM.
Figure 3.
PARP1 KO animals are resistant to PDE6 inhibition induced photoreceptor cell death.
Organotypic retinal explant cultures obtained from wt and PARP1 KO animals were treated with the PDE6 inhibitor zaprinast and compared to untreated wt, rd1 and PARP1 KO cultured retinae. Control wt retinae at P11 in vitro showed minimal immunoreactivity to cGMP-antibody in the ONL (A). Zaprinast treated wt retinae exhibited strongly increased intracellular cGMP levels (B), similar to what was observed in PDE6 mutant rd1 retina (C). While PARP1 KO did not display cGMP immunoreactivity (D), it responded to zaprinast treatment with a moderate, but significant, elevation of cGMP positive ONL cells (E). Accumulation of PAR as an indirect marker for PARP activity (F–J) was found in large amounts only in zaprinast treated wt or rd1 ONL but notably absent in PARP1 KO preparations and wt retinae. The TUNEL assay for dying cells generally followed a similar pattern (K–O): The rd1 mutant or PDE6 inhibition resulted in marked increases of positive cells in the ONL only. The bar graph (P) illustrates the quantification of the three parameters assayed. Explant cultures from n = 3–6 animals were used for each treatment situation and genotype. Error bars represent SEM.